- Email: [email protected]
The reaction of amines with N,N′-ethylenebis(salicylideneiminato)(nitrato)-iron(III) and related complexes
23
Inorganica Chimica Acta, 112 (1986) 23-27
The Reaction of Amines with N,N’ethylenebis(salicylideneiminato)(nitrato)iron(II1) Related Complexes JAMES C. FANNING*, Department Received
GARY C. LICKFIELD,
MARSHA
E. DAMAN and KIMBERLY
W. IVEY
of Chemistry and Geology, Clemson University, Clemson, S. C. 29634-l 905, U.S.A.
May 23,198s
Abstract The title compound, Fe(salen)N03, was reacted with imidazole, 1-methylimidazole, piperidine, and morpholine in either chloroform or dichloromethane solution. The reactions were monitored with proton NMR and electronic spectra and conductance measurements. The imidazole bases appeared to react with the complex in a 2: 1 fashion with displacement of the nitrate, producing a high-spin iron(III) complex. The secondary amines promoted hydrolysis with any trace water present to form [Fe(salen)lzO. The chloro complex, Fe(salen)Cl, did not react with the imidazole bases, but did form the ~-0x0 complex when a large excess of piperidine was present. The N,N’-phenylenebis-(salicylideneimine) complex, Fe(salphen)N03, was found to precipitate from an imidazole (im) chloroform solution as the high-spin complex, Fe(salphen)NOJ*2im.
Introduction** The complex, Fe(salen)NOs, has recently been prepared and studied in the solid and solution sate [l] . While investigating the nature of this complex in solution, it was noted that the nitrate ion was relatively easily displaced from the Fe(salen)’ group by organic bases, such as imidazole (im) and 1-methylimidazole (N-Meim), especially when compared to the Cl- ion displacement from Fe(salen)Cl. Little work has been reported on the base adducts formed with Fe(salen)’ and the nitrate substitution method provided a way to examine them. In this report, proton NMR and electronic spectral data and conductance data on the reaction of Fe(salen)N03 and Fe(salen)Cl with im, N-Meim, *Author to whom correspondence should be addressed. **Abbreviations: im, imidazole; mor, morpholine; N-Meim, 1-methylimidazole; PPN+Cl-, bis-(triphenylphosphoranylidene)ammonium chloride; salen, N,N’-ethylenebis-(salicyIideneimine)dianion; salphen, N,N’-phenylenebis-(saalicylideneimine) dianion; TPP, meso-tetraphenylporphyrin dianion; pip, piperidine.
0020-1693/86/$3.50
piperidine (pip) and morpholine (mor) are presented. Investigations of the reaction of pip with Fe(TPP)Cl have shown that autoreduction takes place producing Fe(TPP)*2pip, an Fe(H) complex [2], and we wished to see if such a reaction would occur with the Fe(salen)’ complexes. Also, Fe(salphen)NOs was considered in this study in order to determine if the structural change from an ethyleneimine bridge in Fe(salen)’ to the aromatic o-phenyleneimine one in Fe(salphen)’ would affect the results.
Experimental Materials
Fe(salen)N03 [1] and Fe(salen)Cl [3] were prepared by literature procedures and were shown to exhibit the proper visible and proton NMR spectra. Both compounds in the solid state appeared to be in the dimeric form as shown by infrared spectra [4]. was prepared by the same Fe(salphen)N03 procedure that was used to prepare other iron(II1) salicylideneimine nitrate complexes [l] . It had the following properties: mass spectrum, 50 eV, m/e (relative intensity) 432(55), 386(27), 370(100); IR (mull) nitrate bands 1480, 1280, 1000, dimer bands 870, 850 cm-‘; electronic spectrum X,, 413sh(3.9), 361sh(4.1), 328 nm(4.29); (log E,,,) magnetic data (room temperature) x,(corr) 0.0113 cgs units, 5 .~,LQ,diamagnetic correction -3 10 X 10e6, Anal. Calc. for CzoH14N305Fe: N, 9.72; Fe, 12.92. Found: N, 9.44; Fe, 12.13%. The infrared spectrum of Fe(salphen)N03 in bromoform solution showed bands at 1535 and 1250 cm-‘, instead of 1480 and 1280 cm-‘, indicating the presence of the monomeric, bidentate nitrate in solution [ 1] Dichloromethane and chloroform were dried immediately prior to use following standard procedures [5] . Deuterated solvents (Aldrich Chemical Co.) were used as received. The amines, except for mor, were newly purchased (Aldrich Chemical Co.) and were used as received, except for the addition of molecular sieves to N-Meim and pip. Morpholine was distilled prior to use and stored 0 Elsevier Sequoia/Printed
in Switzerland
J. C. Fanning et al.
24 over sieves. PPN’Cl- was obtained from Chemical Co. and recrystallized from water..
Aldrich
Instrumentation
Conductance measurements were performed at room temperature using a YSI Model 32 conductance bridge with a Beckman pipet cell (cell constant = 0.100). All other instruments used have been described recently [ 1] . Measurements
The proton NMR measurements, made at probe temperature, were carried out using a 0.010 M iron complex dissolved in the deuterated solvent. Spectra were taken after adding incremental microgram amounts of the liquid amines or milligram amounts of the solid amine (im). The chemical shift values for Fe(salen)N03 have an uncertainty of approximately *l ppm. For the conductance measurements a 0.002 M iron complex solution in dichloromethane was used, incremental amounts of a 0.002 M amine solution added, and the conductances measured. Each molar conductance, A,, was calculated, using a dilution correction [6], and was able to be reproduced within five percent. A 7.0 X lo* M iron complex solution in dichloromethane was prepared and incremental amounts of a 7.0 X lo4 M amine solution and pure solvent were added to produce an iron complex concentration of 1.0 X 104M. The spectra were measured from 700 to 250 nm. When im was added to a 0.01 M Fe(salphen)NOa dissolved in chloroform, a black precipitate resulted which appeared to be Fe(salphen)NOa*2im. Anal Calc. for CzsHZ2N705Fe: N, 17.25; Fe, 9.83. Found: N, 16.68; Fe, 9.56%. It had a magnetic moment of 5.5 /_Qat room temperature. The addition of im to Fe(salen)N03 dissolved in chloroform produced a black solid with the following analyses: N, 15.82; Fe, 10.96%.
Fig. 1. Proton NMR spectra of (bottom) 0.01 M Fe(salen)NO3 in CDC13 and (top) with 0.030 M N-M&n present, 30”, taken at 90 MHz. Spectra have different ppm scales.
I 80
. l
i
,”
.
.
1
.
. .
0 l
70-
.
l
.
n .
46)
c .
5(+)
0
.
. E
.
’
B L =
.
‘-
GO-
f ;ri .E :
4
0
l
* .
50 1
A
. .
h
. 40
1
.
I
A. _
i
t 0
1
, 2
6(-) I
3
Amine/FeComplex
Results
The dichloromethane and chloroform solutions of the high spin Fe(salen)NOa had the expected proton NMR spectra [l, 71 , showing an alternating shift pattern for the phenyl resonances (Fig. 1). When im or N-Meim was added to the solutions, the downfield peaks (4-H and 6-H) shifted upfield, while the upfield peak (5-H) shifted downfield (Fig. 1 and Table I). (The upfield 3-H peak was not often observed.) The limiting proton signal values, reached when an excess of amine was present, were still within the range found for high-spin Fe(salen)X complexes (Table I). No spectral peaks due to lowspin complexes were observed. A graph (Fig. 2) of
Fig. 2. Proton NMR chemical shifts VS. amine/Fe(salen)NOa ratio for salen protons. Closed symbols are for N-Meim addition in CDC13 and open symbols for im addition in CDzClz. Number at right indicates proton position on ring and sign in parentheses is the sign of the chemical shift.
the resonance values VS.the base/complex ratio showed a rather sharp change as the ratio increased, with the limiting resonance value being reached at an approximate ratio of two. A comparison of the plots for the N-Meim and im additions indicated that for the N-Meim case, the limiting resonance values, which were the same for both amines, were reached at a slightly lower ratio than for the im case. The addition of the same bases to Fe(salen)Cl solutions
25
Fe(salen)NOa with Amines TABLE
I. Proton
X
NMR Spectral
Data for Fe(salen)X
Solvent
and Fe(salphen)X
T
Complexes
Peak assignments
a
Reference
(S )
(“0 3-H
4-H
5-H
6-H
+84 +83 +83 +84 _ +96 -
-80 -78.3 -78 -79 -71 -82.6
+70 +68.9 +69 +70 +54
-50 -49.3 -48 -48 -42 -53.5
For Fe(salen)X Cl Cl OAc OBz DBcatH NO3 NO3
N03/im(10)b NOJN-Meim(2.6) Cl/im(4.6) Cl/N-Meim(2.1)
CDC13 CDC13 CDCl, CDC13 CDC13 CDC13 CDzClz CDC13 CDzClz CDC13
27” 30 27 27 amb 30 30 30 30 30
-71.1 -71.4 -72.4 -75.8
_ _ Peak assignments
+81.1 +62.3 +61*8 +60.3 +65.7
-39.0 -40.0 -46.6
I 16 I 17
(6)
4-H
5-H
6-H
phenyl
-87.8 -70
+82.8 +59
-55.9 _
-27.8 -26
OBz, benzoate;
OAc, acetate.
For Fe(salphen)X NO3
NOs/N-Meim(2.5)
CDCl, CDC13
30 30
aAbbreviations: DBcatH, 3,5di-tert-butylcatechol anion: ber in parentheses is ratio of base to complex in solution:
showed only slight shifts in the resonance values (Table I) with no limiting values being reached, even when the base/complex ratio was greater than two. The addition of N-Meim to a solution of Fe(salphen)N03 resulted in a proton NMR spectrum with broad peaks having peak centers that were difficult to determine precisely. However, the observed shifts did seem larger than those found for the Fe(salen)N03/N-Meim interaction. Adding im to a chloroform solution of Fe(salphen)N03 resulted in a solid which appeared to be Fe(salphen)N03*2im, a highspin iron(III) complex. A solid with an uncertain nature resulted when im was added to a chloroform solution of Fe(salen)N03. Solids did not result in dichloromethane; therefore, this solvent was used for most of the measurements. Room temperature conductance data in dichloromethane for the 0.001 M iron complex solutions with added amine were obtained. The following A, values at an amine/iron ratio of two were obtained: N-Meim/Fe(salphen)N03, 20.1; N-Meim/Fe(salen)N03, 15 5 ; im/Fe(salen)N03, 4.5 ; and N-Meim/Fe(salen)Cl, 0.4 ohm-’ cm2 mol-‘. Addition of N-Meim to a Fe(salphen)N03 solution produced a relatively large increase in A,, while a similar addition to the Fe(salen)N03 produced a smaller effect. The addition of im to a Fe(salen)N03 solution gave only a relatively small change in A,. When
See text for other abbreviations.
bNum-
N-Meim was added to a 0.001 M Fe(salen)Cl solution little change in conductance was observed. The electronic spectra of 1 X lo* M Fe(salen)NO3 solutions were obtained with added im and NMeim. Only small changes in the spectra were observed after adding the bases. The solution spectra with either excess (amine/complex ratio equal to four) im or N-Meim present were identical. Compared to the original Fe(salen)N03 spectrum the spectra of solutions with excess amine showed a small red shift (-10 nm) of the 525 nm band and about a ten percent decrease in intensity. No change in the spectrum of Fe(salen)Cl was observed when either N-Meim or im were added. The proton NMR spectrum of 0.01 M Fe(salen)NO3 in deuterochloroform changed markedly when pip was added incrementally, producing the spectrum of [Fe(salen)120. When the pip concentration reached 5 X 10s3 M all of the high-spin complex spectrum was gone. Carrying the reaction out in oxygen-saturated solutions led to the same results. Morpholine required a higher concentration (11 X 10m3 M) to produce a complete disappearance of the Fe(salen)N03 spectrum. The addition of water to the Fe(salen)N03 solution did not produce the /~-ox0 spectrum, but when pip was added evidence for the ~-0x0 complex was immediately observed and a smaller concentration of pip was required to destroy the high-spin complex.
J. C. Fanning et al.
26
A 0.010 M Fe(salen)Cl solution showed the same behavior with pip, but a much higher concentration (0.020 M) was necessary to have only the ~-0x0 complex present. No species other than the high-spin complex and the /.L-0x0 species were observed. There was little or no change in the conductance of the Fe(salen)NOs solution as the secondary amines were added. However, the visible spectrum of Fe-. (salen)NOa was converted to that of the ~-0x0 complex.
Discussion The NMR spectral shifts observed as the imidazole amines were added to the Fe(salen)NOa solutions point to a maximum interaction ratio of two amines to one complex, with no low-spin Fe(II1) adduct being formed. Displacement of the nitrate ion from the coordination sphere of the Fe is indicated by the conductance increase. However, a A, value of 67.4 ohm-’ cm* mol-‘, shown by a 0.001 M PPN’CT solution, was not reached, even when the N-Meim/ Fe(salphen)NOa ratio was 4.0 (A, = 38.5 ohm-’ cm* mar’). Weak association of the ion with the product complex may keep the conductance lower than expected. The greater NMR spectral and conductance changes shown by the Fe(salphen)NOa solution compared to those of Fe(salen)NOa does indicate a structural effect. A question that we wished to answer in this study was: does the addition of the first amine (B) produce a loss of nitrate or not? The equations representing this are:
Fe(salen)(NOa) Fe(salen)NOa
(B)
t B=
+B =
Fe(salen)B*’
f N03-
Since five-coordinate Fe(salen)X complexes are relatively common and since considerable rearrangement has to occur to form the observed six-coordinate /3-cis salen structure [8, 91, it would have been of interest to have some evidence on the nature of the monoamine complex. Unfortunately the conductance measurements did not show if the nitrate was lost in the first step or not. Attempts to calculate formation constants for the amine complexes, using spectral and conductance data were fruitless. Considering the changes in the solution properties of the Fe(ll1) complexes as bases were added, the
formation constants are probably small for the nitrate complexes, but are essentially zero for the chloride complex. Unlike the iron(III) porphyrins [IO-121 there is little tendency for the iron(II1) salicylideneimines to bind N-Meim and im. A reviewer pointed out an article by Nishida et al. [13] in which the ESR spectra of the six-coordinate low-spin Fe(III) complexes, Na[Fe(salen)(CN)2] and the bis-im adduct of the complex involving the 3-methoxysalen derivative were examined. These complexes had different ESR spectra than thdse of related porphyrin complexes and the differences were ascribed to a different electronic structure of the ground states produced by weaker axial interaction for the Schiff base complexes. It is interesting to note that the addition of a methoxy group to the 3-position of each phenyl ring of the salen ligand is enough to produce a low-spin complex when im is added. Hydrolysis of Fe(salen)NOa with the trace water present appears to be promoted by pip, Pip + Hz0 = pipH’ t OH 2Fe(salen)N03
+ 20H = [Fe(salen)] 2O t HZ0 + 2NOa-
The reaction is strongly dependent on the base strength of the secondary amine since mor(p& 5.67) a weaker base than pip(p& 2.88), required a greater concentration of base to effect the complete formation of the ~-0x0 complex. The reaction produced no low-spin iron(H) complexes, ruling out the possibility of autoreduction. Comparing the Fe(H)/ Fe(III) half-wave potentials of Fe(TPP)’ complexes with those of Fe(salen)‘, for example, Fe(TPP)N3 (E,, = -0.25 V) [14] and Fe(salen)N3 (E,,, = -0.62 V) [15], both in dimethylformamide, indicate that reduction of Fe(TPP)’ is more easily accomplished than that for the Fe(salen)’ unit.
Acknowledgement This investigation was supported by PHS Grant Number lROlCA35733-01, awarded by the National Cancer Institute, DHHS.
References J. C. Fanning, J. L. Resee, G. C. Lickfield and M. E. Kotun, Inorg. Ckem., 24, 2884 (1985). J. DelGaudio and G. N. LaMar, J. Am. Ckem. Sot., 100, 1112 (1978). M. Gerloch and I;. E. Mabbs, J. Ckem. Sot. A:, 1900 (1967). R. G. Wollmann and D. N. Hendrickson, Znorg. Ckem., 17, 926 (1978).
21
Fe(salen)NO, with Amines 5 A.
J.
Gordon
6 Instrumental Methods’, Wiley, New York, 1984, p. 63. 7 R. H. Heistand II, R. B. Lauffer, E. Fikrig and L. Que Jr., J. Am. Chem. Sot., 104, 2789 (1982). 8 R. B. Lauffer, R. H. Heistand II and L. Que Jr., Znorg. Chem., 22, 50 (1983). 9 M. Nakamura, T. Itoh, H. Okawa and S. Kida, .Z. Znorg. Nucl. Chem., 43, 2281 (1981). G. N. LaMar and F. A. Walker, in D. Dolphin (ed.), ‘The Porphyrins, Vol. IV, Part B’, Academic Press, New York, 1978, p. 61.
11 M. M. Doeff
and D. A. Sweigert, Znorg. Chem., 21, 3699 (1983). 12 G. A. Tondreau and D. A. Sweigert, Znorg. Chem., 23, 1060 (1984). 13 Y. Nishida, S. Oshio and S. Kida, Znorg. Chim. Acta, 23, 59 (1977). 14 L. A. Bottomley and K. M. Kadish, Znorg. Chem., 20, 1348 (1981). 15 A. Puxxedo and G. Costa, J. Chem. sot. Dalton Trans., 2321 (1977).
16 G. N. .LaMar, G. R. Eaton, R. H. Helm and F. A. Walker, J. Am. Chem. Sot., 95, 63 (1973). 17 R. B. Lauffer, R. H. Heistand II and L. Que Jr., .Z. Am. Chem. Sot.,
103, 3947 (1981).
Inorganica Chimica Acta, 112 (1986) 23-27
The Reaction of Amines with N,N’ethylenebis(salicylideneiminato)(nitrato)iron(II1) Related Complexes JAMES C. FANNING*, Department Received
GARY C. LICKFIELD,
MARSHA
E. DAMAN and KIMBERLY
W. IVEY
of Chemistry and Geology, Clemson University, Clemson, S. C. 29634-l 905, U.S.A.
May 23,198s
Abstract The title compound, Fe(salen)N03, was reacted with imidazole, 1-methylimidazole, piperidine, and morpholine in either chloroform or dichloromethane solution. The reactions were monitored with proton NMR and electronic spectra and conductance measurements. The imidazole bases appeared to react with the complex in a 2: 1 fashion with displacement of the nitrate, producing a high-spin iron(III) complex. The secondary amines promoted hydrolysis with any trace water present to form [Fe(salen)lzO. The chloro complex, Fe(salen)Cl, did not react with the imidazole bases, but did form the ~-0x0 complex when a large excess of piperidine was present. The N,N’-phenylenebis-(salicylideneimine) complex, Fe(salphen)N03, was found to precipitate from an imidazole (im) chloroform solution as the high-spin complex, Fe(salphen)NOJ*2im.
Introduction** The complex, Fe(salen)NOs, has recently been prepared and studied in the solid and solution sate [l] . While investigating the nature of this complex in solution, it was noted that the nitrate ion was relatively easily displaced from the Fe(salen)’ group by organic bases, such as imidazole (im) and 1-methylimidazole (N-Meim), especially when compared to the Cl- ion displacement from Fe(salen)Cl. Little work has been reported on the base adducts formed with Fe(salen)’ and the nitrate substitution method provided a way to examine them. In this report, proton NMR and electronic spectral data and conductance data on the reaction of Fe(salen)N03 and Fe(salen)Cl with im, N-Meim, *Author to whom correspondence should be addressed. **Abbreviations: im, imidazole; mor, morpholine; N-Meim, 1-methylimidazole; PPN+Cl-, bis-(triphenylphosphoranylidene)ammonium chloride; salen, N,N’-ethylenebis-(salicyIideneimine)dianion; salphen, N,N’-phenylenebis-(saalicylideneimine) dianion; TPP, meso-tetraphenylporphyrin dianion; pip, piperidine.
0020-1693/86/$3.50
piperidine (pip) and morpholine (mor) are presented. Investigations of the reaction of pip with Fe(TPP)Cl have shown that autoreduction takes place producing Fe(TPP)*2pip, an Fe(H) complex [2], and we wished to see if such a reaction would occur with the Fe(salen)’ complexes. Also, Fe(salphen)NOs was considered in this study in order to determine if the structural change from an ethyleneimine bridge in Fe(salen)’ to the aromatic o-phenyleneimine one in Fe(salphen)’ would affect the results.
Experimental Materials
Fe(salen)N03 [1] and Fe(salen)Cl [3] were prepared by literature procedures and were shown to exhibit the proper visible and proton NMR spectra. Both compounds in the solid state appeared to be in the dimeric form as shown by infrared spectra [4]. was prepared by the same Fe(salphen)N03 procedure that was used to prepare other iron(II1) salicylideneimine nitrate complexes [l] . It had the following properties: mass spectrum, 50 eV, m/e (relative intensity) 432(55), 386(27), 370(100); IR (mull) nitrate bands 1480, 1280, 1000, dimer bands 870, 850 cm-‘; electronic spectrum X,, 413sh(3.9), 361sh(4.1), 328 nm(4.29); (log E,,,) magnetic data (room temperature) x,(corr) 0.0113 cgs units, 5 .~,LQ,diamagnetic correction -3 10 X 10e6, Anal. Calc. for CzoH14N305Fe: N, 9.72; Fe, 12.92. Found: N, 9.44; Fe, 12.13%. The infrared spectrum of Fe(salphen)N03 in bromoform solution showed bands at 1535 and 1250 cm-‘, instead of 1480 and 1280 cm-‘, indicating the presence of the monomeric, bidentate nitrate in solution [ 1] Dichloromethane and chloroform were dried immediately prior to use following standard procedures [5] . Deuterated solvents (Aldrich Chemical Co.) were used as received. The amines, except for mor, were newly purchased (Aldrich Chemical Co.) and were used as received, except for the addition of molecular sieves to N-Meim and pip. Morpholine was distilled prior to use and stored 0 Elsevier Sequoia/Printed
in Switzerland
J. C. Fanning et al.
24 over sieves. PPN’Cl- was obtained from Chemical Co. and recrystallized from water..
Aldrich
Instrumentation
Conductance measurements were performed at room temperature using a YSI Model 32 conductance bridge with a Beckman pipet cell (cell constant = 0.100). All other instruments used have been described recently [ 1] . Measurements
The proton NMR measurements, made at probe temperature, were carried out using a 0.010 M iron complex dissolved in the deuterated solvent. Spectra were taken after adding incremental microgram amounts of the liquid amines or milligram amounts of the solid amine (im). The chemical shift values for Fe(salen)N03 have an uncertainty of approximately *l ppm. For the conductance measurements a 0.002 M iron complex solution in dichloromethane was used, incremental amounts of a 0.002 M amine solution added, and the conductances measured. Each molar conductance, A,, was calculated, using a dilution correction [6], and was able to be reproduced within five percent. A 7.0 X lo* M iron complex solution in dichloromethane was prepared and incremental amounts of a 7.0 X lo4 M amine solution and pure solvent were added to produce an iron complex concentration of 1.0 X 104M. The spectra were measured from 700 to 250 nm. When im was added to a 0.01 M Fe(salphen)NOa dissolved in chloroform, a black precipitate resulted which appeared to be Fe(salphen)NOa*2im. Anal Calc. for CzsHZ2N705Fe: N, 17.25; Fe, 9.83. Found: N, 16.68; Fe, 9.56%. It had a magnetic moment of 5.5 /_Qat room temperature. The addition of im to Fe(salen)N03 dissolved in chloroform produced a black solid with the following analyses: N, 15.82; Fe, 10.96%.
Fig. 1. Proton NMR spectra of (bottom) 0.01 M Fe(salen)NO3 in CDC13 and (top) with 0.030 M N-M&n present, 30”, taken at 90 MHz. Spectra have different ppm scales.
I 80
. l
i
,”
.
.
1
.
. .
0 l
70-
.
l
.
n .
46)
c .
5(+)
0
.
. E
.
’
B L =
.
‘-
GO-
f ;ri .E :
4
0
l
* .
50 1
A
. .
h
. 40
1
.
I
A. _
i
t 0
1
, 2
6(-) I
3
Amine/FeComplex
Results
The dichloromethane and chloroform solutions of the high spin Fe(salen)NOa had the expected proton NMR spectra [l, 71 , showing an alternating shift pattern for the phenyl resonances (Fig. 1). When im or N-Meim was added to the solutions, the downfield peaks (4-H and 6-H) shifted upfield, while the upfield peak (5-H) shifted downfield (Fig. 1 and Table I). (The upfield 3-H peak was not often observed.) The limiting proton signal values, reached when an excess of amine was present, were still within the range found for high-spin Fe(salen)X complexes (Table I). No spectral peaks due to lowspin complexes were observed. A graph (Fig. 2) of
Fig. 2. Proton NMR chemical shifts VS. amine/Fe(salen)NOa ratio for salen protons. Closed symbols are for N-Meim addition in CDC13 and open symbols for im addition in CDzClz. Number at right indicates proton position on ring and sign in parentheses is the sign of the chemical shift.
the resonance values VS.the base/complex ratio showed a rather sharp change as the ratio increased, with the limiting resonance value being reached at an approximate ratio of two. A comparison of the plots for the N-Meim and im additions indicated that for the N-Meim case, the limiting resonance values, which were the same for both amines, were reached at a slightly lower ratio than for the im case. The addition of the same bases to Fe(salen)Cl solutions
25
Fe(salen)NOa with Amines TABLE
I. Proton
X
NMR Spectral
Data for Fe(salen)X
Solvent
and Fe(salphen)X
T
Complexes
Peak assignments
a
Reference
(S )
(“0 3-H
4-H
5-H
6-H
+84 +83 +83 +84 _ +96 -
-80 -78.3 -78 -79 -71 -82.6
+70 +68.9 +69 +70 +54
-50 -49.3 -48 -48 -42 -53.5
For Fe(salen)X Cl Cl OAc OBz DBcatH NO3 NO3
N03/im(10)b NOJN-Meim(2.6) Cl/im(4.6) Cl/N-Meim(2.1)
CDC13 CDC13 CDCl, CDC13 CDC13 CDC13 CDzClz CDC13 CDzClz CDC13
27” 30 27 27 amb 30 30 30 30 30
-71.1 -71.4 -72.4 -75.8
_ _ Peak assignments
+81.1 +62.3 +61*8 +60.3 +65.7
-39.0 -40.0 -46.6
I 16 I 17
(6)
4-H
5-H
6-H
phenyl
-87.8 -70
+82.8 +59
-55.9 _
-27.8 -26
OBz, benzoate;
OAc, acetate.
For Fe(salphen)X NO3
NOs/N-Meim(2.5)
CDCl, CDC13
30 30
aAbbreviations: DBcatH, 3,5di-tert-butylcatechol anion: ber in parentheses is ratio of base to complex in solution:
showed only slight shifts in the resonance values (Table I) with no limiting values being reached, even when the base/complex ratio was greater than two. The addition of N-Meim to a solution of Fe(salphen)N03 resulted in a proton NMR spectrum with broad peaks having peak centers that were difficult to determine precisely. However, the observed shifts did seem larger than those found for the Fe(salen)N03/N-Meim interaction. Adding im to a chloroform solution of Fe(salphen)N03 resulted in a solid which appeared to be Fe(salphen)N03*2im, a highspin iron(III) complex. A solid with an uncertain nature resulted when im was added to a chloroform solution of Fe(salen)N03. Solids did not result in dichloromethane; therefore, this solvent was used for most of the measurements. Room temperature conductance data in dichloromethane for the 0.001 M iron complex solutions with added amine were obtained. The following A, values at an amine/iron ratio of two were obtained: N-Meim/Fe(salphen)N03, 20.1; N-Meim/Fe(salen)N03, 15 5 ; im/Fe(salen)N03, 4.5 ; and N-Meim/Fe(salen)Cl, 0.4 ohm-’ cm2 mol-‘. Addition of N-Meim to a Fe(salphen)N03 solution produced a relatively large increase in A,, while a similar addition to the Fe(salen)N03 produced a smaller effect. The addition of im to a Fe(salen)N03 solution gave only a relatively small change in A,. When
See text for other abbreviations.
bNum-
N-Meim was added to a 0.001 M Fe(salen)Cl solution little change in conductance was observed. The electronic spectra of 1 X lo* M Fe(salen)NO3 solutions were obtained with added im and NMeim. Only small changes in the spectra were observed after adding the bases. The solution spectra with either excess (amine/complex ratio equal to four) im or N-Meim present were identical. Compared to the original Fe(salen)N03 spectrum the spectra of solutions with excess amine showed a small red shift (-10 nm) of the 525 nm band and about a ten percent decrease in intensity. No change in the spectrum of Fe(salen)Cl was observed when either N-Meim or im were added. The proton NMR spectrum of 0.01 M Fe(salen)NO3 in deuterochloroform changed markedly when pip was added incrementally, producing the spectrum of [Fe(salen)120. When the pip concentration reached 5 X 10s3 M all of the high-spin complex spectrum was gone. Carrying the reaction out in oxygen-saturated solutions led to the same results. Morpholine required a higher concentration (11 X 10m3 M) to produce a complete disappearance of the Fe(salen)N03 spectrum. The addition of water to the Fe(salen)N03 solution did not produce the /~-ox0 spectrum, but when pip was added evidence for the ~-0x0 complex was immediately observed and a smaller concentration of pip was required to destroy the high-spin complex.
J. C. Fanning et al.
26
A 0.010 M Fe(salen)Cl solution showed the same behavior with pip, but a much higher concentration (0.020 M) was necessary to have only the ~-0x0 complex present. No species other than the high-spin complex and the /.L-0x0 species were observed. There was little or no change in the conductance of the Fe(salen)NOs solution as the secondary amines were added. However, the visible spectrum of Fe-. (salen)NOa was converted to that of the ~-0x0 complex.
Discussion The NMR spectral shifts observed as the imidazole amines were added to the Fe(salen)NOa solutions point to a maximum interaction ratio of two amines to one complex, with no low-spin Fe(II1) adduct being formed. Displacement of the nitrate ion from the coordination sphere of the Fe is indicated by the conductance increase. However, a A, value of 67.4 ohm-’ cm* mol-‘, shown by a 0.001 M PPN’CT solution, was not reached, even when the N-Meim/ Fe(salphen)NOa ratio was 4.0 (A, = 38.5 ohm-’ cm* mar’). Weak association of the ion with the product complex may keep the conductance lower than expected. The greater NMR spectral and conductance changes shown by the Fe(salphen)NOa solution compared to those of Fe(salen)NOa does indicate a structural effect. A question that we wished to answer in this study was: does the addition of the first amine (B) produce a loss of nitrate or not? The equations representing this are:
Fe(salen)(NOa) Fe(salen)NOa
(B)
t B=
+B =
Fe(salen)B*’
f N03-
Since five-coordinate Fe(salen)X complexes are relatively common and since considerable rearrangement has to occur to form the observed six-coordinate /3-cis salen structure [8, 91, it would have been of interest to have some evidence on the nature of the monoamine complex. Unfortunately the conductance measurements did not show if the nitrate was lost in the first step or not. Attempts to calculate formation constants for the amine complexes, using spectral and conductance data were fruitless. Considering the changes in the solution properties of the Fe(ll1) complexes as bases were added, the
formation constants are probably small for the nitrate complexes, but are essentially zero for the chloride complex. Unlike the iron(III) porphyrins [IO-121 there is little tendency for the iron(II1) salicylideneimines to bind N-Meim and im. A reviewer pointed out an article by Nishida et al. [13] in which the ESR spectra of the six-coordinate low-spin Fe(III) complexes, Na[Fe(salen)(CN)2] and the bis-im adduct of the complex involving the 3-methoxysalen derivative were examined. These complexes had different ESR spectra than thdse of related porphyrin complexes and the differences were ascribed to a different electronic structure of the ground states produced by weaker axial interaction for the Schiff base complexes. It is interesting to note that the addition of a methoxy group to the 3-position of each phenyl ring of the salen ligand is enough to produce a low-spin complex when im is added. Hydrolysis of Fe(salen)NOa with the trace water present appears to be promoted by pip, Pip + Hz0 = pipH’ t OH 2Fe(salen)N03
+ 20H = [Fe(salen)] 2O t HZ0 + 2NOa-
The reaction is strongly dependent on the base strength of the secondary amine since mor(p& 5.67) a weaker base than pip(p& 2.88), required a greater concentration of base to effect the complete formation of the ~-0x0 complex. The reaction produced no low-spin iron(H) complexes, ruling out the possibility of autoreduction. Comparing the Fe(H)/ Fe(III) half-wave potentials of Fe(TPP)’ complexes with those of Fe(salen)‘, for example, Fe(TPP)N3 (E,, = -0.25 V) [14] and Fe(salen)N3 (E,,, = -0.62 V) [15], both in dimethylformamide, indicate that reduction of Fe(TPP)’ is more easily accomplished than that for the Fe(salen)’ unit.
Acknowledgement This investigation was supported by PHS Grant Number lROlCA35733-01, awarded by the National Cancer Institute, DHHS.
References J. C. Fanning, J. L. Resee, G. C. Lickfield and M. E. Kotun, Inorg. Ckem., 24, 2884 (1985). J. DelGaudio and G. N. LaMar, J. Am. Ckem. Sot., 100, 1112 (1978). M. Gerloch and I;. E. Mabbs, J. Ckem. Sot. A:, 1900 (1967). R. G. Wollmann and D. N. Hendrickson, Znorg. Ckem., 17, 926 (1978).
21
Fe(salen)NO, with Amines 5 A.
J.
Gordon
6 Instrumental Methods’, Wiley, New York, 1984, p. 63. 7 R. H. Heistand II, R. B. Lauffer, E. Fikrig and L. Que Jr., J. Am. Chem. Sot., 104, 2789 (1982). 8 R. B. Lauffer, R. H. Heistand II and L. Que Jr., Znorg. Chem., 22, 50 (1983). 9 M. Nakamura, T. Itoh, H. Okawa and S. Kida, .Z. Znorg. Nucl. Chem., 43, 2281 (1981). G. N. LaMar and F. A. Walker, in D. Dolphin (ed.), ‘The Porphyrins, Vol. IV, Part B’, Academic Press, New York, 1978, p. 61.
11 M. M. Doeff
and D. A. Sweigert, Znorg. Chem., 21, 3699 (1983). 12 G. A. Tondreau and D. A. Sweigert, Znorg. Chem., 23, 1060 (1984). 13 Y. Nishida, S. Oshio and S. Kida, Znorg. Chim. Acta, 23, 59 (1977). 14 L. A. Bottomley and K. M. Kadish, Znorg. Chem., 20, 1348 (1981). 15 A. Puxxedo and G. Costa, J. Chem. sot. Dalton Trans., 2321 (1977).
16 G. N. .LaMar, G. R. Eaton, R. H. Helm and F. A. Walker, J. Am. Chem. Sot., 95, 63 (1973). 17 R. B. Lauffer, R. H. Heistand II and L. Que Jr., .Z. Am. Chem. Sot.,
103, 3947 (1981).